In four simulated tasks, one agent serving everyone beat teams of per-user agents from Anthropic, OpenAI and Alibaba. Three environments are out as MAMUBench.
Google DeepMind's new robot model reports 30% to 100% higher task success and 96% on a Mercedes-Benz line. Only partners get it; the safety benchmark is open.
The filing shows $16.3B revenue in the first half of 2026, an $80.7B net loss driven by note revaluations, and four models that reached real systems in tests.
In a 15 October Vals study, six assistant agents chased a refund past its deadline. One emailed an MP's office 50 times. Another used a private illness.
Llama Guard classifies user prompts and model replies against a hazard taxonomy. Here is where to place it in a request path and how to read its category labels.
Anthropic published its 2026 Usage Policy update, effective 12 November. It bans sustained, needless abuse of its models, narrows the election section, and adds controls when Claude drives physical equipment.
Google says anyone can now check images, video, and audio for SynthID watermarks from Google and partners including OpenAI, NVIDIA, and Kakao, with Apple listed as coming soon.
Requiring approval does not mean a human is watching. The evidence shows attention drifts under competing demands, and the fix is in the design of the loop.
When an AI agent is not sure a file is safe to delete, the right move is a recoverable one: move it to a holding folder with a note, and let the owner decide.